REDUCING GREENHOUSE GAS EMISSIONS FROM LONG-DISTANCE ...

REDUCING GREENHOUSE GAS EMISSIONS FROM LONG-DISTANCE TRAVEL 1

BUSINESS: HOW FAR CAN WE GO? 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

44 45

Ruohan Li

Graduate Research Assistant

The University of Texas at Austin

[email protected]

Kara M. Kockelman

(Corresponding author)

Professor and Dewitt Greer Fellow

Department of Civil, Architectural and Environmental Engineering


[email protected]

512-471-0210

Jooyong Lee

Graduate Research Assistant


[email protected]

Transportation Research Record, 2021 https://doi.org/10.1177%2F03611981211036682

ABSTRACT

Long-distance (LD) travels account for over 30-percent of person-trip miles, with energy and

emissions impacts for everyone. Long-distance business travel can often be handled remotely, so

targeting such trips for cost, time, and emissions savings may be a wise strategy for protection of climate,

budgets, and human health. To better appreciate Americans’ LD travel choices, a 73-question online

survey was conducted in 2019 that captured 2,327 long-distance (over 200 mileseach way) trips made by

929 respondents over the prior 12 months, of which 490 round-trips were for business purposes.

Predictive models for LD trips per adult per year, overnights, LD travel times, and willingness to

participate remotely and/or pay carbon taxes for those trips were developed using only respondents in

Austin. As expected, those with higher education tend to travel more often - for both business and non-

business purposes, everthing else constant. Persons who travel long-distance more frequently are more

likely to spend more nights at their destinations, while spending less time in transit/en route. Those

from households with more workers and those with fewer children at home tend to be more receptive

to having their employers pay carbon taxes. Some level of viability is found in approximately a

quarter of the 298 long-distance business trips made by Austinites, and the respondents are willing to

remotely participate in 44% of them. In other words, Austinites are willing to remotely participate in

slightly over 10% of their business trips overall. This is definitely not enough to address carbon

emissions in the LD travel sector.

Keywords: Long-Distance Travel, Business Travel, Carbon Tax, Greenhouse Gases, Remote Participation,

Travel Behaviors

INTRODUCTION

1 46

mailto:[email protected]

maizyjeong

Highlight

2

Though relatively infrequent, long-distance (LD) travels account for a significant portion of transportaton’s 1 vehicle-miles traveled (VMT) and greenhouse gases (GHG) (Georggi and Pendyala, 2000). According to 2 the U.S. 2009 National Household Travel Survey (USDOT, 2009), 75% of all person-trips made in the US 3 were less than 10 miles (one way), accounting for 28.9% of VMT, while trips over 100 miles (one way) 4 made up less than 1% of all vehicle trips but contributed 15.5% of household-based VMT (Schiffer, 2012, 5 Llorca et al., 2018). Looking at all personal travel (including airline, bus and train travel), long-distance 6 travel delivers around 30% of U.S. person-miles (Aultman-Hall, 2018). Average trip lengths and the 7 number of LD trips per person are rising over time in other countries as well (Dargay and Clark, 2012). 8 And in Europe, long-distance trips (counted as those over 100 km or 62.1 mi) constitute about 50% of 9 person-miles (Rich and Mabit, 2011). 10

According to Chester (2008), the GHG life-cycle impact contributions per PMT of different planes are shown 11 in Table 1. 12 13 Table 1. GHG Life-Cycle Impact Contribution by Type of Aircraft 14

Type of Aircraft GHG (g/PMT)

Embraer 145 290

Boeing 737 210

Boeing 747 320

Air travel is experiencing an increasing trend not only in the United States, but around the world as well. 15 Chi and Baek (2013) found that US air passenger-miles increased 52% for international and 32% for 16 domestic flights between 1996 and 2010, while Mayor and Tol (2010) predicted that from 2000 to 2025, 17 an average annual growth rate of 6.8% is to be observed in passenger kilometers worldwide, and a 5.2% 18 from 2000 to 2050. 19

Trip purposes of tourism (or leisure) and business are key to understanding LD travel (Koenker, 2003; 20 Millán et al., 2016). Travel & tourism GDP summed to $4,740 billion globally in 2015, with 23.4% of that 21 for business reasons (Turner & Freiermuth, 2016). In the US, year-215 business travel was 28.4% of the 22 US travel and tourism total (Statista, 2016). In general, the goal of leisure trips is to be there, at one’s 23 destination. Physical participation is typically fundamental to tourism and leisure, including visiting family 24 members or friends. In contrast, business trips are typically designed to exchange information, 25 professionally network, and/or conduct deals. These goals can regularly be achieved by other means, thanks 26 to conference call technologies and advances in high-speed internet and audio-visual optioins. Video calls 27 allow us to communicate effectively with high definition with almost no time delay. Remote participation 28 requires significantly less time, energy, and resources than business travel, so it is the focus of this study, 29 in terms of reducing LD travel while saving energy, emissions, and other resources. 30

Travel is a major cost for businesses and their workers workwide. For example, the World Travel and 31 Tourism Conference (WTTC, 2015) estimates that U.S. business travel’s contribution to GDP is over $200 32 billion annually, making it 30% of all U.S. travel and tourism expenditures. On average, U.S. companies 33 spend $949 per worker pursuing a domestic trip (Certify, 2019). LD travel, represented in large part by air 34 travel, is responsible for a disproportionate share of environmental impacts. Airline mode dominates for 35 person-trips over 500 miles, and accounts for more than 90% of mode share at travel distances over 1,000 36 miles (Moeckel et al., 2015.) Airlines’ energy and emissions implications per person-mile are typically 37 higher than those of ground travel, with each commercial airline’s seat-mile responsible for roughly the 38 same amount of GHG emissions (over 400 gm per mile) as a mid-size sedan, which typically carries 2 39 persons (on average) during long-distance trip-making. (Chester and Horvath, 2009; Hoyer et al., 2001; 40 Ottelin et al., 2014). By filling all seats in most light-duty passenger vehicles, one can reduce one’s carbon 41 footprint by 75%, versus flying. By using a small sedan (rather than a large SUV or pickup truck, for 42

3

example) with hybrid drivetrain or battery power, GHG emissions fall even further, under almost all 1 settings. Right-sizing and sharing of electric self-driving vehicles may reduce those numbers further (Lee 2 and Kockelman 2019; Loeb et al., 2018; Liu et al., 2017; Michalek et al., 2018; Nichols et al., 2015; Reiter 3 and Kockelman, 2017), but widespread adoption of such technologies is simply not possible in the near 4 term. In order to reduce the environmental, cost and other impacts of LD passenger travels in the near and 5 long terms, it is important to have a deeper understanding of what motivates such travel. 6

LITERATURE REVIEW 7

Key variables impacting LD trip frequency, distance, mode choice, and destination include a traveler’s 8 household income (Van Wee et al., 2006; Sandow & Westin, 2010), age (Collia et al., 2003), education 9 level (Holz-Rau et al., 2014), presence of children in the household (LaMondia et al., 2016a), and so forth 10 (Cho, 2013). Specific events and objectives – like professional conferences and vacation opportunities, 11 friends’ weddings and family funerals, tennis tournaments and music concerts (Yang et al., 2016; Burke & 12 Woolcock, 2013; McKercher et al., 2008; Aguilera, 2008) regularly motivate LD travel for many different 13 types of people. 14

Regional models of statewide and nationwide trip-making have sought to evaluate the LD travel impacts 15 for regions and nations (Erhardt et al., 2007; Rohr et al., 2013; Bernardin Jr et al., 2017; LaMondia et al. 16 2016b; Perrine et al., 2017) and can be used to plan the transportation network required in a certain region. 17 Models of LD travelers’ mode choices were also developed to estimate the modes used for LD travels and 18 the decision-making process of LD travelers (Scheiner, 2010; Reichert & Holz-Rau, 2015.) LD travels 19 made at different times of day (e.g. day-time or overnight) are known to have different characetristics. Thus, 20 LD travels carried out at different times of day were studied as well (Aultman-Hall et al., 2015; Sullivan et 21 al., 2016.) 22

As previously mentioned, tourism and business travels are the two most popular types of LD travels. 23 Tourism and leisure travel occur more frequently in urban areas and in holiday seasons (Große et al., 2019.) 24 Tourism is generally accompanied with aviation and accounted for 8% of global GHG emissions between 25 2009 and 2013 (Lenzen et al., 2018.) Although tourists distinguished their travel to daily commute and 26 considered it occasional, some of them still refused to be responsible to the transport of aircrafts and identify 27 it as a personal responsibility (Becken, 2007.) Therefore, rather than passing the responsibilities to 28 individual tourists, environmental policies and strategies to mitigate the GHG implications from tourism 29 should target the development of the tourism industry to reduce the global environmental impact and 30 decarbonize the industry (Cavallaro et al., 2017; Scott et al., 2016; Brendehaug et al., 2017; Gössling et al., 31 2017). 32

Another type of LD travel, business travel, is more economically motivated and industrialized than tourism, 33 which is more leisure-based. Business travel usually consists of visiting certain business sites, meeting with 34 business counterparts, or contracting to business activities. With economic growth, the significance of 35 business travel will become more important than ever before (Kulendran & Wilson, 2000; Baker et al., 36 2015.) However, it is possible that travelers on LD business travel might be not as motivated to be 37 environmentally concious as non-business LD travelers. 38

One of the factors that affects LD business travelers’ behavior is that their travel expenses can be reimbursed 39 from their employers (Schaeffer, 2009; Cai et al., 2011); thus, they might be less aware of their implications 40 on society and the environment. Moreover, frequent LD business travelers are usually highly educated, 41 have higher income, status and position in their organization; thus, it is not straightforward to manage and 42 control their behavior (Gustafson, 2012). An in-depth interview revealed that some frequent business 43 travelers even consider their experience as a passenger as “moments of relaxation” or “time off,” although 44 they identify their activities and destination as job-related (Unger et al., 2016). The characteristics of LD 45 business travels have an inherently aggressive impact on the environment by means of GHG emissions. 46

4

In an effort to reduce the environmental implications and travel costs caused by business LD travels, new 1 communication methods and transport modes have been developed or will be applied in due time. For 2 instance, videoconferencing technologies and remote participation can not only connect people from a long 3 distance without an actual travel but can also allow for communication between groups in different time 4 zones (Julsrud et al., 2012; Pawlak et al., 2015). Proponents of video-conferencing note that it can eliminate 5 the necessity of business air travel by allowing a face-to-face meeting without physically meeting (Lu & 6 Peeta, 2009; Denstadli et al., 2013). 7

Application of new technologies - including connected and autonomous vehicles, shared autonomous 8 vehicles, and dynamic ride-sharing en route can help reduce the energy intensity of LD travel, while keeping 9 departure times more flexible, travel costs low, and access direct between origins and destinations 10 (LaMondia et al., 2016b; Perrine et al., 2017). But large-scale adoption of self-driving vehicles will not be 11 reasonable until the 2030s and 2040s or 2050s in mode locations (Quarles, 2018), and and range limitations 12 on all-electric vehicles (He et al., 2019; Loeb et al., 2018) makes it difficult to be a feasible option for LD 13 travel in short term. 14

DATA DESCRIPTION 15

In March 2019, an online survey was conducted to collect information on American adults’ “long-distance” 16 (over 200 miles, each-way) travel behaviors, opinions, and demographics, including numbers of LD 17 business and non-business trips taken over the prior 12 months. LD trip purposes, durations, origins and 18 destinations, cost and other attributes were recorded, as shown in Figures 1 through 3. The initial data set 19 captured 2,327 round-trip LD trips (total) made by 929 respondents (age 18 years or older), with 490 (21%) 20 trips for business purposes. Due to sample advertising via largely Austin-based and UT Austin Parking 21 Services distribution channels, the majority of data came from residents of the Austin, Texas region and 22 those with parking passes at the University of Texas at Austin. To ensure more reliable estimates with 23 robust sample expansion, only data provided by Austin area residents are used in this paper, resulting in n 24 = 1,599 LD trips made by 623 respondents, out of which 298 (18.6%) were for business purposes. Using 25 either sample (n = 929 or n = 623 respondents), the average number of LD trips per respondent per year is 26 about 2.5. While Austinites tend to be more educated than the the average American, they appear similar 27 in LD trip-making frequency. 28

To correct for sampling biases within the Austin region, the 2017 American Community Survey (ACS) 29 splits by household income, gender, age, marital status, education, and ehtnicity (for adults only) were 30 compared to sample demographics, as shown in Figs. 1(a) through 1(f), and then used to weight and thus 31 correct for all values presented elsewhere in this paper, including model estimation for parameter 32 predictions and other results. 33

5

1

Figure 1. Demographics in Sampled Data versus Austin’s Adult Population 2

As shown in Fig. 1, survey respondents tend to be of higher income, younger, more often female, single, 3 highly educated, and Caucasian than the average Austin-region adult. Their households’ average vehicle 4 ownership is 1.66, which is 16% lower than the national average of 1.97 vehicles per household, according 5 to the 2017 National Household Travel Survey (FHWA, 2018). Females are clearly over-represented in the 6 sample (at 67.8% of respondents vs. 49.8% of the Austin population). To overcome such sampling bias, all 7 survey responses are weighted here by using the American Community Survey (ACS) Public Use 8

6

Microdata Sample (PUMS) demographic shares for the Austin area. The weights were obtained by 1 subcategorizing both the PUMS data and the survey sample into 288 groups, across 6 age classes, 2 genders, 2 6 educational attainment levels, and 4 marital status classifications. 3

4

(a) Annual Long-Distance Round-Trip Counts by Trip Purpose 5

6

7

(b) Long-Distance Trip Departure by Travel Purpose 8

Figure 2. Long-Distance (>200 mile each way) Round-Trip Counts by Business/Non-Business 9 Purposes (using unweighted Austin sample) 10

Figure 2(a) shows weighted LD trip fractions for business and non-business trips across 623 respondents, 11 with 55.2% of respondents reporting no LD business trips (over 200 miles [each way]) in the 12 months 12 prior to the survey and 1.09% reporting no non-business LD trips over the past 12 months. Among all 5,057 13

0%

10%

20%

30%

40%

50%

60%

Frac

tio

n

Annual Number of Trips

Business Non-Business

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

Frac

tio

n

Month of Trip Departure

Business

Non-Business

7

LD trips reported by Austinites, 77.9% are non-business purposes, while 22.1% are for business purposes. 1 To save on GHG emissions and LD travel costs, it would be useful to forego non-business trips, at at least 2 shorten their distances and lower their per-mile carbon footprints, but technology for remote participation 3 is not an option for almost all personal trips. 4

In terms of trip-making seasonality, Georggi and Pendyala (2000) noted how business trip counts fall 5 considerably between October and December, while Mallett (1999) found US non-business LD travel 6 peaking over the Thanksgiving (late November), Christmas (late December), and New Year’s holidays, 7 thus lagging the drop in business trip-making by about 1 month. 8

Figure 2(b) shows month of departure for 1,599 LD trips made by Austinites, with higher fractions 9 ofbusiness trips (versus non-business LD trips) made in February, March, January and November. 10 December alone claims over one-fifth of all non-business trips, while February claims over a quarter of the 11 year’s LD business trips. Consecutive months that claim nearly half of all reported LD business-trip 12 departures are January, February, and March, versus December, January, and February for Austinites’ non-13 business departures. 14

The average reimbursement share for the data set’s 489 business trips is 72%. Interestingly, 19% of LD 15 business trips received zero reimbursement, while 44% were fully reimbursed. Roughly one-third of all 16 these LD business trips are less than 300 miles in one-way, Euclidean (straight-line) distance, with travelers 17 spending an average of 6.2 days and 5.2 nights on each, with very high variability (coefficents of variation 18 of almost 4), as shown in Tables 2 and 3. 19

20

Table 2. Hours Spent by Different Modes during Long-Distance Business Trips 21

Variable #Obs. Mean Std. Dev. Min Max Median

Hours Stopped to Destination 287 1.111 hr 4.169 hr 0 hr 72 hr 0.25 hr

Hours by Air to Destination 287 3.196 3.581 0 45 3

Hours Drive Alone to

Destination 284 0.774 1.264 0 9 0

Hours Carpool to Destination 288 0.737 2.417 0 19 0

Hours by TNC* to Destination 287 0.326 1.533 0 30 0

Hours by Train to Destination 287 0.065 0.270 0 2 0

Hours by Bus to Destination 287 0.105 0.516 0 5 0

Hours by Walk to Destination 287 0.049 0.362 0 8 0

Hours by Other Mode(s) to

Destination (including airport

shuttles, recreational vehicles,

and rental cars)

288 0.097 0.981 0 18 0

Hours Stopped on Return Trip 275 1.216 3.016 0 24 0.05

Hours by Air on Return Trip 275 3.015 3.582 0 45 3

Hours Drive Alone on Return

Trip 273 0.963 2.079 0 17 0

Hours Carpool on Return Trip 275 0.510 1.426 0 13 0

Hours by TNC* on Return

Trip 274 0.399 1.854 0 30 0

Hours by Train on Return Trip 275 0.049 0.224 0 2 0

Hours by Bus on Return Trip 275 0.117 0.551 0 5 0

Hours Walk on Return Trip 275 0.020 0.093 0 1 0

8

Hours by Other Mode on

Return Trip (Including airport

shuttle, recreational vehicle,

and rental car)

275 0.095 1.017 0 19 0

*Note: TNC = ride-hailing trips (like Lyft & Uber) 1

Table 3. LD Business Travel Cost Breakdown 2

Variable #Obs. Mean Std. Dev. Min Max Median

Food Expenses 277 $ 174.83 $ 266.29 $ 0 $ 6000 $ 120

Accomodation 278 518.32 680.32 0 6300 300

Administration 276 191.18 530.66 0 4000 0

Visa/Passport 278 6.32 43.06 0 360 0

Transportation 274 354.19 681.06 0 12,000 200

Gas 278 16.43 41.12 0 671 0

Other Cost 278 24.76 246.81 0 3000 0

Reimbursement 268 1023.3 1253.9 0 12,460 720

3

To further understand business trip attributes, travel purposes were asked of all respondents, and multiple 4 choices were allowed. Figure 3 shows those survey results, with “networking” and “fulfillment of job 5 duties” being the top 2 reasons given. Under the response option of “Other”, respondents often mentioned 6 the need to perform work-related experiments and tests, rehearse and/or perform sports or music, and pursue 7 legal activities (including depositions and meetings with lawyers and/or judges). 8

9

Figure 3. Trip Purpose Distribution (n=1070) 10

Unfortunately, just 24.9% of Austinites’ LD business trips are reported as possibly substitutable via remote 11 participation technologies, with more detailed classification of these 106 responses shown in Table 4. The 12 great majority (75.1%) were deemed “impossible”, in terms of remote substitution. Among 383 trips, 114 13 could not be done remotely because the respondents wanted to travel to the destination area apart from their 14 primary trip purpose. Among these 114 trips, 85 indicated that meeting in person was needed to help close 15 a business deal, while 71 cited the lack of infrastructure or technology limitations. Among the other 208 (of 16 the 383 “impossible to substitute”) trips, many respondents noted their involving a transfer of property or 17 other activities (like testing in person) that required their physical presence. 18

0.00%

5.00%

10.00%

15.00%

20.00%

25.00%

30.00%

To network To fulfillspecifications

of my job

To give aspeech,

present, etc.

To improvework

performance

To better thestanding of my

business

To experiencea new

environment

To better mystanding

within thebusiness

Other

Res

po

nse

Rat

e (%

)

9

Table 4. Possibility of vs. Willingness to Remotely Participate in LD Business Trips 1

#Obs. = 106 Willingness

Total Yes No Not sure

Possibility of

Remote

Participation

Yes 29.2% 39.6% 3.32% 72.1%

Not sure 14.9% 12.3% 0.67% 27.9%

Total 44.1% 51.9% 3.98% 100%

2

MODELING RESULTS 3

To appreciate the various factors that make one business traveler answer in one way versus another, 595 4 respondents were first selected, who had experienced long-distance travel (over 200 miles each way) within 5 the last 12 months. Among total 623 Austinites sampled in the survey, 595 respondents were used during 6 the model calibration to best fit the model. The excluded 28 respondents did not have the relevant 7 explanatory variable to fit the model. A negative binomial model was used first, to predict the numbers of 8 business and non-business trips per year, with the base case being a married, full-time employed, 18-to-24-9 year-old male with a bachelor’s degree, no children, no pet, an annual household income of $100,000 to 10 $124,999, who has someone that needs care at home and household members over 16 that can’t drive, in a 11 two-member, two-worker household. The results are shown in Tables 5 and 6. 12

Table 5. Parameter Estimates in Prediction of Number of Business Trips per Year (using 13 population-weighted negative binomial count model) 14

Variable Coefficient P-Value

Age 35-44 0.635 0.092

Age 55-64 -0.668 0.038

Household Size = 5 -0.894 0.060

Worker in Household = 1 0.456 0.052

Female -0.492 0.011

Black/African American -2.297 0.003

2 Children in Household 1.734 0.002

Income $30,000 - $49,999 -0.649 0.012

$60,000 to $74,999 -0.723 0.073

$75,000 to $99,999 -1.373 0.000

High School Diploma -1.998 0.004

Some College, no Degree -0.866 0.005

Associate's Degree -2.455 0.017

PhD Degree 0.735 0.002

Employed, working less than 40 hours per week -1.585 0.001

Retired -2.414 0.000

Single -0.809 0.001

Widowed -25.881 0.000

No Household Member Requires Special

Attention 1.471 0.000

No Household Member over 16 w/ Conditions

Prevent Driving 1.459 0.005

10

Constant -1.566 0.007

Dispersion Parameter 2.631 ± 0.367

n = 595, Pseudo R2 = 0.1391 1

The most practically significant variable corresponding to a person’s making more business trips each year 2 was having two children in household, with an incidence rate ratio (IRR) of 5.66 (meaning that the average 3 expected number of trips per year rose by more than a factor of 5 when this binary variable changed from 4 0 to 1), and the second most siginificant one was having a PhD degree, with an IRR of 2.09. Being widowed 5 was by far the most significant variable corresponding to fewer business trips, with an IRR of 5.75 × 10-6 12. Two other key covariates having an associate’s degree (which is about 2 years of college coursework) 7 and being retired, with IRRs of 0.086 and 0.089, respectively, suggesting 91% fewer business trips each 8 year for persons of that type, everything else constant. This result supports the notion that highly educated 9 workers living in larger households are more likely to have more business trips than others. 10

Table 6. Negative Binomial Model Output for Non-Business Trips 11


Age 45-64 -0.319 0.003


3 Workers in Household 0.314 0.040

4 Workers in Household -1.269 0.001

Black/African American -0.319 0.024

Asian/Asian American -0.321 0.014

1 Child in Household 0.316 0.068

Income $75,000 to $99,999 0.251 0.026

Income $150,000 to $199,999 0.417 0.006

Income $200,000 or more 0.538 0.000


Some College, No Degree -0.215 0.083

Employed, working less than 40 hours per week. -0.652 0.005

Student, working part time. 0.639 0.000

Student, not working. 0.750 0.000

No Household Member Requires Special Attention 0.313 0.018

No Household Member over 16 w/ Conditions Prevent Driving 0.543 0.066

Not Sure on Household Member over 16 w/ Conditions Prevent

Driving 0.997 0.008

Constant 0.848 0.005


n = 595, Pseudo R2 = 0.0644 12

The most practically significant variable associated with more non-business trips each year comes with 13 being an unemployed student, with an IRR of 2.12, followed by being a student working part-time, with an 14 IRR of 1.89. Some other important variables (leading to more LD non-business trip-making each year) 15 include having an annual household income higher than the base case and having one child in the household. 16 The most significant variables associated with fewer non-business trips are having only a high school 17 diploma (no higher degrees) and being part-time employed, with IRR values of 0.40 and 0.52, respectively. 18 Having a household size of 3 persons also corresponds to a lower model-predicted number of non-business 19 trips. This result supports that idea that those with fewer household members are more likely to have non-20 business trips than others. 21

11

The number of nights spent at one’s destination of each business trip was predicted by fitting another 1 negative binomial model, as shown in Table 7. Compared to the base case, having an associate’s degree, an 2 annual household income between $20,000 and $39,000, or traveling long-distance for work 5-8 times a 3 year were more likely to be identified with more nights spent at the destination. On the other hand, having 4 a high school diploma, age being over 24, and having 2 children in household were more likely to be 5 identified with less nights spent at the destination. Overall, the more often the individual travels, the more 6 nights he/she is likely to spend at the destination of his/her trips. The respondents were also more likely to 7 spend more nights at their destination if their destination was located farther than 300 miles in straight-line 8 distance from their origin. 9

Table 7. Number of Nights Spent at Business Trip Destinations 10


Annual Long Distance Trips Made 0.011 0.006

Age over 24 -1.564 0.000

2 Children in Household -0.632 0.006

3 Children in Household 0.661 0.000

Household Income $20,000 - $39,999 1.181 0.000


Associate's Degree 1.667 0.000

Master's Degree 0.292 0.048

Student, Working Part-Time -0.570 0.017

Retired 0.708 0.001

Travel Long-Distance for Work 1-4 Times a Year 0.677 0.008

Travel Long-Distance for Work 5-8 Times a Year 0.889 0.000

Trip Longer than 300 Miles Over Land 0.655 0.000

Constant 1.318 0.002


n = 276, Pseudo R2 = 0.1414 11

Finally, the total travel time (in hours) the respondents spent in their travel was predicted by fitting a linear 12 regression model, both to from their destination, the results are shown in Table 8. Unlike the results shown 13 above, the travel time was shorter if the respondent traveled long-distance trips more often. Positive 14 coefficients that appear to be the most practically significant include having more than 5 vehicles, having 15 an annual household income of $10,000 - $19,999 or $30,000 - $39,999, while negative coefficients include 16 having 3 children in household, traveling for work 5 – 8 times per year, and having 1 – 4 vehicles in 17 household. The number of vehicles owned by a household was especially meaningful, by showing that 18 respondents with 1-4 vehicles tend to have shorter total travel time than respondents with no vehicles, while 19 respondents with 5 or more vehicles tend to have longer total travel time than those with no vehicles. 20

Table 8. Linear Regression Model Output for Travel Time (hr) 21


More than 20 Long Distance Trips per Year -2.743 0.048

Household Size = 1 2.695 0.010

1 Worker in Household -1.853 0.078


Annual Income $10,000 - $19,999 6.884 0.010

12

Annual Income $30,000 - $39,999 5.512 0.000

Student, Working Part-Time -3.924 0.000

Single -3.434 0.001

Pet but Do not Need Care While Gone -2.941 0.000

1 - 4 Vehicles -5.800 0.001

More than 5 Vehicles 23.839 0.000

Travel for Work 5-8 Times per Year -6.042 0.000

Constant 18.262 0.000

n = 256, R2 = 0.1009 1

After estimating the model to predict the number of trips made per year, the length of travel, and travel 2 time, each Austinite’s opinion on carbon tax was estimated to understand how much of the travels can be 3 charged for a carbon tax. An ordered logistic model was estimated to predict the respondent’s opinion on 4 charging carbon tax to the companies that require employees to travel longer distance. To avoid carbon tax, 5 companies might reduce unnecessary long-distance trips and replace them to remote participation. In the 6 survey question, five levels of responses included extremely unreasonable, moderately unreasonable, 7 neither reasonable nor unreasonable, moderately reasonable, and extremely reasonable. As shown in Figure 8 4, moderately reasonable gained the highest response rate (40.8%), which implies charging carbon tax to 9 mandatory long-distance travels could be acceptable. The sum of moderately reasonable and extremely 10 reasonable was higher than 60%, which implies that the respondents supports the carbon tax suggested in 11 this paper. 12

13 Figure 4. Distribution of Attitudes towards Carbon Tax (n = 586) 14

According to Table 9, respondents from households with more workers may consider carbon tax to be more 15 reasonable, while those with more children in their household are unlikely to consider carbon tax as a 16 reasonable option. Compared to the base case of having a bachelor’s degree, one with an associate’s degree 17 may also tend to view the carbon tax as less reasonable. This result supports that households with a more 18 productive population favor the carbon tax designed in this paper more so than others. 19

Table 9. Ordinal Logistic Model Output for Charging Carbon Tax 20


0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

Extremelyunreasonable

Moderatelyunreasonable

Neitherreasonable norunreasonable

Moderatelyreasonable

Extremelyreasonable

Frac

tio

n

How Reasonable the Respondent Considers Carbon Tax

13

Household Size = 5 2.640 0.009




Mixed/Multiracial Ethnicity 0.856 0.045

1 Child in Household -1.104 0.041



Associate's Degree -1.315 0.020

No Household Member Requires Special Attention -0.668 0.060

Threshold Std. Dev.

Threshold 1 (1) -2.819 0.364

Threshold 2 (2) -1.789 0.361

Threshold 3 (3) -1.226 0.353

Threshold 4 (4) 0.795 0.348

n = 574, Pseudo R2 = 0.0271 1

CONCLUSIONS 2

This work obtained extensive data on Austinites’ long-distance (over 200-miles each way) trip-3 making for an entire year, and analyzed those data in order to predict number of trips made each year, trip 4 duration, travel time, willingness to pay or have one’s employer pay for carbon offsets on such travel, and 5 willingness to shift to remote participation for business trips. Those making more long-distance business 6 trips tend to reside in larger households, while those making more non-business trips tend to have fewer 7 household members. Those traveling more often tend to spend more nights at each trip’s destination than 8 others. Those with more workers and less children in the household tend to view carbon-taxation of LD 9 travel more reasonable, as a public policy, than others. Unfortunately, just 25% of all LD trip-making 10 appears to be for business purposes, and thus having reasonable potential for remote participation and 11 significant GHG reductions; and (weighted) respondents for just 10% of the LD business trips observed in 12 this data set were willing to substitute their trip using remote participation technology options. Much 13 greater reduction in LD travel’s carbon emissions are needed to achieve climate protection goals equitably 14 in this fast-growing area of GHG emissions. Sharing rides in fully-electric self-driving vehicles, on the 15 ground, rather than in the air, may be key, in addition to LD travelers’ avoiding business-class and first-16 class seating and using carbon taxes on all air travel, ground travel and hotel rentals to purchase offsets 17 for things like rainforest protection, investments in renewable energy and well-insulated homes, battery-18 electric fleets, and reforestation efforts. 19 ACKNOWLEDGEMENTS 20

The authors are grateful for the contribution of Jeffrey Hahm and classmates in CE 392E, Acquisition & 21 Analysis of Transport Data, at the University of Texas at Austin during the Spring 2019 semester for the 22 development of survey questions and data collection. They greatly appreciate the University of Texas at 23 Austin’s Parking & Trasportation Services leaders for distribution of the online survey URL and to Albert 24 Coleman for his editing and submission support. 25

REFERENCES 26 Aguilera, A. (2008). Business travel and mobile workers. Transportation Research Part A: Policy and 27 Practice, 42(8), 1109-1116. 28

14

Aultman-Hall (2018). Incorporating Long-Distance Travel into Transportation Planning in the United 1 States. Institute of Transportation Studies, UC Davis. Accessed on July 23, 2019 at: 2 https://escholarship.org/uc/item/0ft8b3b5 3

Aultman-Hall, L., Harvey, C., LaMondia, J. J., & Ritter, C. (2015). Design and response quality in a one-4 year longitudinal survey of overnight and long-distance travel. Transportation Research Procedia, 11, 5 136-153. 6

Baker, D., Merkert, R., & Kamruzzaman, M. (2015). Regional aviation and economic growth: 7 cointegration and causality analysis in Australia. Journal of Transport Geography, 43, 140-150. 8

Becken, S. (2007). Tourists' perception of international air travel's impact on the global climate and 9 potential climate change policies. Journal of Sustainable Tourism, 15(4), 351-368. 10

Bernardin Jr, V. L., Ferdous, N., Sadrsadat, H., Trevino, S., & Chen, C. C. (2017). Integration of National 11 Long-Distance Passenger Travel Demand Model with Tennessee Statewide Model and Calibration to Big 12 Data. Transportation Research Record, 2653(1), 75-81. 13

Boarnet, M., & Crane, R. (2001). The influence of land use on travel behavior: specification and 14 estimation strategies. Transportation Research Part A: Policy and Practice, 35(9), 823-845. 15

Brendehaug, E., Aall, C., & Dodds, R. (2017). Environmental policy integration as a strategy for 16 sustainable tourism planning: issues in implementation. Journal of Sustainable Tourism, 25(9), 1257-17 1274. 18

Burke, M., & Woolcock, G. (2013). Getting to the game: travel to sports stadia in the era of transit-19 oriented development. In Australian Sport (pp. 64-83). Routledge. 20

Cai, H., Fang, H., & Xu, L. C. (2011). Eat, drink, firms, government: An investigation of corruption from 21 the entertainment and travel costs of Chinese firms. The Journal of Law and Economics, 54(1), 55-78. 22

Cavallaro, F., Galati, O. I., & Nocera, S. (2017). Policy strategies for the mitigation of GHG emissions 23 caused by the mass-tourism mobility in coastal areas. Transportation Research Procedia, 27, 317-324. 24

Certify. (2019). Understanding the average cost of business travel. Accessed on July 23, 2019 at: 25 https://www.certify.com/Infographic-TheAverageCostOfBusinessTravel.aspx 26

Chester, M. V. (2008). Life-cycle environmental inventory of passenger transportation in the United 27 States. 28

Chester, M. V., & Horvath, A. (2009). Environmental assessment of passenger transportation should 29 include infrastructure and supply chains. Environmental Research Letters, 4(2), 024008. 30

Chi, J., & Baek, J. (2013). Dynamic relationship between air transport demand and economic growth in 31 the United States: A new look. Transport Policy, 29, 257-260. 32

Cho, H. D. (2013). The factors that affect long-distance travel mode choice decisions and their 33 implications for transportation policy. Ph.D. Dissertation, University of Florida. 34

Collia, D. V., Sharp, J., & Giesbrecht, L. (2003). The 2001 national household travel survey: A look into 35 the travel patterns of older Americans. Journal of Safety Research, 34(4), 461-470. 36

Dargay, J. M., & Clark, S. (2012). The determinants of long distance travel in Great Britain. 37 Transportation Research Part A: Policy and Practice, 46(3), 576-587. 38

15

Denstadli, J. M., Gripsrud, M., Hjorthol, R., & Julsrud, T. E. (2013). Videoconferencing and business air 1 travel: Do new technologies produce new interaction patterns? Transportation Research Part C: 2 Emerging Technologies, 29, 1-13. 3

Erhardt, G. D., Freedman, J., Stryker, A., Fujioka, H., & Anderson, R. (2007). Ohio long-distance travel 4 model. Transportation Research Record, 2003(1), 130-138. 5

Federal Highway Administration (FWHA, 2018). 2017 NHTS Data User Guide. Washington D.C. USA, 6 Retrieved from https://nhts.ornl.gov/assets/2017UsersGuide.pdf 7

Georggi, N. L., & Pendyala, R. M. (2000). An analysis of long-distance travel behavior of the elderly and 8 the low-income (Master's thesis, University of South Florida). 9

Gössling, S., Fichert, F., & Forsyth, P. (2017). Subsidies in aviation. Sustainability, 9(8), 1295. 10

Große, J., Fertner, C., & Carstensen, T. A. (2019). Compensatory leisure travel? The role of urban 11 structure and lifestyle in weekend and holiday trips in Greater Copenhagen. Case Studies on Transport 12 Policy, 7(1), 108-117. 13

Gustafson, P. (2012). Managing business travel: Developments and dilemmas in corporate travel 14 management. Tourism Management, 33(2), 276-284. 15

Holz-Rau, C., Scheiner, J., & Sicks, K. (2014). Travel distances in daily travel and long-distance travel: 16 what role is played by urban form?. Environment and Planning A, 46(2), 488-507. 17

He, Y., Kockelman, K. M., & Perrine, K. A. (2019). Optimal locations of US fast charging stations for 18 long-distance trip completion by battery electric vehicles. Journal of cleaner production, 214, 452-461. 19

Hoyer, K. G., & Næss, P. (2001). Conference tourism: a problem for the environment, as well as for 20 research?. Journal of Sustainable Tourism, 9(6), 451-470. 21

Julsrud, T. E., Hjorthol, R., & Denstadli, J. M. (2012). Business meetings: do new videoconferencing 22 technologies change communication patterns?. Journal of Transport Geography, 24, 396-403. 23

Koenker, D. P. (2003). Travel to work, travel to play: on Russian tourism, travel, and leisure. Slavic 24 Review, 62(4), 657-665. 25

Kulendran, N., & Wilson, K. (2000). Is there a relationship between international trade and international 26 travel?. Applied Economics, 32(8), 1001-1009. 27

LaMondia, J. J., Moore, M., & Aultman-Hall, L. (2016a). Matching Household Life-Cycle Characteristics 28 to Clustered Annual Schedules of Long-Distance and Overnight Travel. Transportation Research Record, 29 2594(1), 11-17. 30

LaMondia, J. J., Fagnant, D. J., Qu, H., Barrett, J., & Kockelman, K. (2016b). Shifts in long-distance 31 travel mode due to automated vehicles: Statewide mode-shift simulation experiment and travel survey 32 analysis. Transportation Research Record, 2566(1), 1-11. 33

Lenzen, M., Sun, Y. Y., Faturay, F., Ting, Y. P., Geschke, A., & Malik, A. (2018). The carbon footprint 34 of global tourism. Nature Climate Change, 8(6), 522. 35

Liu, J., Kockelman, K. M., Boesch, P. M., & Ciari, F. (2017). Tracking a system of shared autonomous 36 vehicles across the Austin, Texas network using agent-based simulation. Transportation, 44(6), 1261-37 1278. 38

16

Llorca, C., Molloy, J., Ji, J., & Moeckel, R. (2018). Estimation of a Long-Distance Travel Demand Model 1 using Trip Surveys, Location-Based Big Data, and Trip Planning Services. Transportation Research 2 Record, 2672(47), 103-113. 3

Loeb, B., Kockelman, K. M., & Liu, J. (2018). Shared autonomous electric vehicle (SAEV) operations 4 across the Austin, Texas network with charging infrastructure decisions. Transportation Research Part C: 5 Emerging Technologies, 89, 222-233. 6

Lu, J. L., & Peeta, S. (2009). Analysis of the factors that influence the relationship between business air 7 travel and videoconferencing. Transportation Research Part A: Policy and Practice, 43(8), 709-721. 8

Mallett, W. J. (1999). Long-distance travel by women: results from the 1995 American Travel Survey. 9 Transportation Research Record, 1693(1), 71-78. 10

Mayor, K., & Tol, R. S. (2010). Scenarios of carbon dioxide emissions from aviation. Global 11 Environmental Change, 20(1), 65-73. 12

McKercher, B., Chan, A., & Lam, C. (2008). The impact of distance on international tourist movements. 13 Journal of Travel Research, 47(2), 208-224. 14

Michalek, J. J., Chester, M., Jaramillo, P., Samaras, C., Shiau, C. S. N., & Lave, L. B. (2011). Valuation 15 of plug-in vehicle life-cycle air emissions and oil displacement benefits. Proceedings of the National 16 Academy of Sciences, 108(40), 16554-16558. 17

Millán, A., Fanjul, M. L., & Moital, M. (2016). Segmenting the business traveler based on emotions, 18 satisfaction, and behavioral intention. Psychology & Marketing, 33(2), 82-93. 19

Moeckel, R., Fussell, R., & Donnelly, R. (2015). Mode choice modeling for long-distance travel. 20 Transportation Letters, 7(1), 35-46. 21

Nichols, B. G., Kockelman, K. M., & Reiter, M. (2015). Air quality impacts of electric vehicle adoption 22 in Texas. Transportation Research Part D: Transport and Environment, 34, 208-218. 23

Ottelin, J., Heinonen, J., & Junnila, S. (2014). Greenhouse gas emissions from flying can offset the gain 24 from reduced driving in dense urban areas. Journal of Transport Geography, 41, 1-9. 25

Pawlak, J., Polak, J. W., & Sivakumar, A. (2015). Towards a microeconomic framework for modelling 26 the joint choice of activity–travel behaviour and ICT use. Transportation Research Part A: Policy and 27 Practice, 76, 92-112. 28

Perrine, K. A., Kockelman, K. M., & Huang, Y. (2017). Anticipating long-distance travel shifts due to 29 self-driving vehicles. In 97th Annual Meeting of the Transportation Research Board, and under review for 30 publication in Transport Policy. 31

Quarles, N. T. (2018). Americans’ plans for acquiring and using electric, shared, and self-driving vehicles 32 and costs and benefits of electrifying and automating US bus fleets (Master’s Thesis, The University of 33 Texas at Austin, Austin, Texas). Retrieved from https://repositories.lib.utexas.edu/handle/2152/64110. 34

Reichert, A., & Holz-Rau, C. (2015). Mode use in long-distance travel. Journal of Transport and Land 35 Use, 8(2), 87-105. 36

Reiter, M. S., & Kockelman, K. M. (2017). Emissions and exposure costs of electric versus conventional 37 vehicles: A case study in Texas. International Journal of Sustainable Transportation, 11(7), 486-492. 38

Rich, J., & Mabit, S. L. (2012). A long-distance travel demand model for Europe. European Journal of 39 Transport and Infrastructure Research, 12(1), 1-20. 40

https://repositories.lib.utexas.edu/handle/2152/64110

17

Rohr, C., Fox, J., Daly, A., Patruni, B., Patil, S., & Tsang, F. (2013). Modeling long-distance travel in 1 great britain. Transportation Research Record, 2344(1), 144-151. 2

Sandow, E., & Westin, K. (2010). The persevering commuter–Duration of long-distance commuting. 3 Transportation Research Part A: Policy and Practice, 44(6), 433-445. 4

Schaeffer, M. S. (2009). Putting expense fraud under the microscope. Financial Executive, 25(10), 42-45. 5

Scheiner, J. (2010). Interrelations between travel mode choice and trip distance: trends in Germany 1976–6 2002. Journal of Transport Geography, 18(1), 75-84. 7

Schiffer, R. G. (2012). Long-distance and rural travel transferable parameters for statewide travel 8 forecasting models (Vol. 735). Transportation Research Board. 9

Scott, D., Gössling, S., Hall, C. M., & Peeters, P. (2016). Can tourism be part of the decarbonized global 10 economy? The costs and risks of alternate carbon reduction policy pathways. Journal of Sustainable 11 Tourism, 24(1), 52-72. 12

Statista (2016). Business travel in the U.S.. Accessed on July 23, 2019 at: 13 https://www.statista.com/study/21059/business-travel-in-the-us-statista-dossier/ 14

Sullivan, J., LaMondia, J. J., Harvey, C., Garrison, C., & Aultman-Hall, L. (2016). Analysis of the Long-15 Distance and Overnight Travel Tour Planning Process. Transportation Research Record, 2594(1), 1-10. 16

Turner, R., & Freiermuth, E. (2016). Travel & Tourism Economic Impact 2016 World. Council of the 17 Federation. 18

Unger, O., Uriely, N., & Fuchs, G. (2016). The business travel experience. Annals of Tourism Research, 19 61, 142-156. 20

US Census Bureau (2018). Current Population Survey, Annual Social and Economic Supplement. 21 Washington, D.C. USA. 22

US Department of Transportation. (USDOT, 2009). National household travel survey. 23

Van Wee, B., Rietveld, P., & Meurs, H. (2006). Is average daily travel time expenditure constant? In 24 search of explanations for an increase in average travel time. Journal of Transport Geography, 14(2), 25 109-122. 26

World Travel & Tourism Council. (WTTC, 2015). Economic Impact: United States. Accessed on July 23, 27 2019 at: https://www.wttc.org/economic-impact/country-analysis/country-data/#undefined 28

Yang, L., Shen, Q., & Li, Z. (2016). Comparing travel mode and trip chain choices between holidays and 29 weekdays. Transportation Research Part A: Policy and Practice, 91, 273-285.30

REDUCING GREENHOUSE GAS EMISSIONS FROM LONG-DISTANCE ...

Documents

Transcript of REDUCING GREENHOUSE GAS EMISSIONS FROM LONG-DISTANCE ...